11.13 Shell Functions

Nowadays, it is difficult to find a shell that does not support shell functions at all. However, some differences should be expected.

When declaring a shell function, you must include whitespace between the ‘)’ after the function name and the start of the compound expression, to avoid upsetting ksh. While it is possible to use any compound command, most scripts use ‘{...}’.

     $ /bin/sh -c 'a(){ echo hi;}; a'
     hi
     $ ksh -c 'a(){ echo hi;}; a'
     ksh: syntax error at line 1: `}' unexpected
     $ ksh -c 'a() { echo hi;}; a'
     hi

Inside a shell function, you should not rely on the error status of a subshell if the last command of that subshell was exit or trap, as this triggers bugs in zsh 4.x; while Autoconf tries to find a shell that does not exhibit the bug, zsh might be the only shell present on the user's machine.

Likewise, the state of ‘$?’ is not reliable when entering a shell function. This has the effect that using a function as the first command in a trap handler can cause problems.

     $ bash -c 'foo() { echo $?; }; trap foo 0; (exit 2); exit 2'; echo $?
     2
     2
     $ ash -c 'foo() { echo $?; }; trap foo 0; (exit 2); exit 2'; echo $?
     0
     2

DJGPP bash 2.04 has a bug in that return from a shell function which also used a command substitution causes a segmentation fault. To work around the issue, you can use return from a subshell, or ‘AS_SET_STATUS’ as last command in the execution flow of the function (see Common Shell Constructs).

Not all shells treat shell functions as simple commands impacted by ‘set -e’, for example with Solaris 10 /bin/sh:

     $ bash -c 'f() { return 1; }; set -e; f; echo oops'
     $ /bin/sh -c 'f() { return 1; }; set -e; f; echo oops'
     oops

Shell variables and functions may share the same namespace, for example with Solaris 10 /bin/sh:

     $ f () { :; }; f=; f
     f: not found

For this reason, Autoconf (actually M4sh, see Programming in M4sh) uses the prefix ‘as_fn_’ for its functions.

Handling of positional parameters and shell options varies among shells. For example, Korn shells reset and restore trace output (‘set -x’) and other options upon function entry and exit. Inside a function, IRIX sh sets ‘$0’ to the function name.

It is not portable to pass temporary environment variables to shell functions. Solaris /bin/sh does not see the variable. Meanwhile, not all shells follow the Posix rule that the assignment must affect the current environment in the same manner as special built-ins.

     $ /bin/sh -c 'func() { echo $a;}; a=1 func; echo $a'
     ⇒
     ⇒
     $ ash -c 'func() { echo $a;}; a=1 func; echo $a'
     ⇒1
     ⇒
     $ bash -c 'set -o posix; func() { echo $a;}; a=1 func; echo $a'
     ⇒1
     ⇒1

Some ancient Bourne shell variants with function support did not reset ‘$i, i >= 0’, upon function exit, so effectively the arguments of the script were lost after the first function invocation. It is probably not worth worrying about these shells any more.

With AIX sh, a trap on 0 installed in a shell function triggers at function exit rather than at script exit. See Limitations of Shell Builtins.